The Metamorphoses of Kinship. Maurice Godelier
and listening to what people have to say about themselves and others, about their past and their present, and trying to confront discourse with actual practice. To be sure, these discourses and practices must be placed in a much longer time frame than that of an individual’s personal memories and references: this time frame is that of the modern history of European societies.
Such an attitude entails combining various approaches and methods from the social sciences, first among which are those used by historians, who try to bring to life a past more often unknown than forgotten or invented, and those used by anthropologists, whose profession demands long immersion in a contemporary society and its observation, at a remove, as it were, but also from within. What does the anthropologist have to say about this evolution?
Let us imagine someone who knows little of the latest developments in anthropology but who is knowledgeable about the social sciences and is now seeking quickly to discover what has become of kinship studies. This reader would likely start her investigation with the formerly widely accepted opinion that the study of kinship ‘is to anthropology what logic is to philosophy or the nude is to art; it is the basic discipline of the subject’. Despite the somewhat dubious comparisons, this formulation by Robin Fox, author of a still-useful book on kinship,10 seemed at the time of its publication (1961) to state a long-established given.
Without necessarily returning to the founding fathers of anthropology (in particular L. H. Morgan, who in 1871 published his huge Systems of Consanguinity and Affinity of the Human Family11), the mere mention of a few of the great names in the discipline – Pitt-Rivers, Kroeber, Radcliffe-Brown, Evans-Pritchard, Fortes, Murdock, Lévi-Strauss, Lounsbury, Dumont, Needham – all of whom owe something of their renown to their contribution to kinship studies, should be enough to reassure the non-specialist that kinship is indeed an area in which anthropology excels and an object whose study is more or less its specialty.
Our reader would therefore probably be astonished, as she scrolls through the various available sources on the computer screen, to discover that the study of kinship has practically disappeared from the course lists of numerous American university anthropology departments as well as a certain number in Europe who have followed suit.
In the space of forty years, kinship – which seemed to have come out rather well in the many tough battles through which competing generations of anthropologists sought to define or redefine the object, its principles and (biological and/or social) foundations – had finally dissolved of its own accord. It had become a non-object for many anthropologists themselves, even before today’s self-styled ‘postmodernists’ appeared on the scene and set about ‘deconstructing’ their discipline. This is indicated by the fact that the leading figures of this movement – Marcus, Fisher, Clifford, et al. – made practically no mention of kinship in the many books in which they draw up a critical inventory of anthropology and propose new objects of study for its reconstruction.12
In reality, as the rest of the present work intends to show, this apparent absence stems from the fact that, far from having vanished, the object ‘kinship’ has emigrated to other areas of anthropology where it is being refashioned and linked to new questions. In other words, the analysis of kinship has simply deserted those places where anthropology had been running in circles for decades, bogged down in insoluble problems by false principles. The blanks left by this desertion are not necessarily a sign that the announced death has occurred.
But let us begin at the beginning.
MORGAN, THE FOUNDER
Why begin with the American Lewis Henry Morgan? Because he epitomizes the contradictions facing anthropology from the start. At the same time, he also shows the conditions under which fieldwork and the interpretations anthropologists propose of what they have observed can slowly acquire a scientific character and constitute a new type of knowledge of the other and oneself, one that no longer merely projects onto this other the prejudices of the anthropologist and his or her culture, garbed in discourse borrowed from the exact sciences.
By way of a reminder, let us recall that, in Morgan’s time (1818–81), the paradigm of scientific explanation was Darwin’s theory of the evolution of species. It was in this context that Morgan – a jurist by profession, a railroad company lawyer in Rochester, New York, and friend and defender of the Indians in their struggle against expropriation and other exactions inflicted by European settlers – became fascinated by Indian customs and decided to devote his life to their study.13
While doing fieldwork among the Seneca, a tribe of the Iroquois confederation, Morgan discovered that their kinship relations displayed a logic of their own that was very different from that of the European and American-European systems. He noted that, where Europeans use two terms to designate the father and the father’s brothers (‘uncles’), the Seneca do not make this distinction and use the same term for these men as well as for all those they classify in the same category with respect to an individual of reference (Ego). He also discovered that, vice versa, where Europeans have a single term ‘cousin’ to designate the children of both father’s and mother’s brothers and sisters, the Seneca have two: one for the children of the father’s brothers and the mother’s sisters, and another for the children of the father’s sisters and the mother’s brothers. In other words, they use different terms to designate the children of their parents’ same-sex and opposite-sex collaterals, distinguishing between what anthropologists have called parallel cousins and cross cousins. And since the terms for parallel cousins of both sexes are the same as those used for brothers and sisters, and since brothers and sisters cannot marry each other, parallel cousins cannot marry each other either, on pain of committing incest. On the other hand, it will often be possible, and even recommended, to marry one’s cross cousin. Finally, owing to the fact that, in contrast to kinship terminologies that classify several individuals under a single term – for instance, the father and all his brothers – European terminologies proceed by describing step by step the relations linking one individual to another, as in the expression ‘my grandfather’s grandfather is my great-great-grandfather’, Morgan concluded that there is a fundamental difference in the principles governing these terminologies. He would call the first ‘classificatory terminologies’ and the second ‘descriptive terminologies’. This opposition would subsequently draw strong criticism, and was later emended.
Morgan also discovered that the composition of the Iroquois exogamous groups could be explained using a descent rule that traced descendants exclusively through the women’s line, whereas in Europe descent is traced through both men and women. He used a Latin word, gens, to designate those groups of individuals who regarded themselves as descending through women from a common female ancestor. This was not a matter of chance. And he called the principle governing these kinship groups the ‘matrilineal descent’ rule. He also noted that, once men married, they left their clan and took up residence with their wife’s people. Lastly, he concluded from these observations that all these elements formed a coherent whole with its own logic, in other words, a ‘system’.
When he extended his study to other North American Indian tribes with different languages and cultures, he discovered that, beyond these differences, a number of them used kinship terminologies that had the same structure as that of the Seneca. This type of structure would come to be called ‘Iroquois’. Other groups, however, such as the Crow and the Omaha, had very different terminologies and marriage rules. Confronted with this diversity but also these convergences, Morgan decided to launch a worldwide survey of kinship terminologies and marriage rules. He drew up a questionnaire describing nearly a hundred possible kin ties with respect to a male or a female Ego, which constituted a sort of family tree ending or starting with Ego, and he sent out nearly one thousand copies to missionaries, civil servants and colonial administrators all over the world.14
Thanks to their replies, Morgan was the first person in history to dispose of such a quantity and diversity of information on kinship practices in societies dispersed widely over the face of the earth. Analysis and comparison of this data showed that the dozens of terminologies collected in totally unrelated languages turned out to be varieties or variants of a few types, which he dubbed Punaluan, Turanian, etc., and which we now, following Murdock, call ‘Hawaiian’, ‘Dravidian’, etc. As a consequence, European kinship systems, too, would appear as varieties of one of